Method for manufacturing a thin film transistor array panel

The present application is a continuation of the U.S. patent application Ser. No. 09/781,987 filed Feb. 14, 2001, which is a divisional of U.S. patent application Ser. No. 09/405,178 filed Sep. 24, 1999, which now became U.S. Pat. No. 6,207,480.

BACKGROUND OF THE INVENTION

(a) Field of the Invention

The present invention relates to a method for manufacturing a thin film transistor (TFT) array panel for a liquid crystal display (LCD).

(b) Description of the Related Art

Thin film transistors used for an LCD have two different types of structure. One is an etch-back type and the other is an etch-stopper type.

When manufacturing. an etch-back type TFT, an amorphous silicon layer and a doped amorphous silicon layer are deposited in sequence and patterned. Next, a metal layer is deposited and patterned to form a data wire including source and drain electrodes, and the doped amorphous silicon layer is etched by using the source and the drain electrodes as etching mask. In the etch stopper type TFT, an etch stopper, which has a large etch selectivity, is formed between an amorphous silicon layer and a doped amorphous silicon layer.

Now, a conventional method for manufacturing a TFT array panel for an LCD will be described with reference to the FIGS. 1A to <b>1</b>D.

FIGS. 1A to <b>1</b>D are cross-sectional views of a TFT array panel, as it undergoes sequential processing steps according to the conventional manufacturing method.

First, as shown in FIG. 1A, an aluminum-neodymium (Al—Nd) layer <b>11</b> and a molybdenum (Mo) layer <b>12</b> are sequentially deposited on a substrate <b>1</b> and patterned to form a gate electrode <b>10</b>. That is, a gate wire having the gate electrode <b>10</b> is formed. A gate insulating layer <b>13</b>, an amorphous silicon layer <b>14</b> and an n+ amorphous silicon layer <b>15</b> are sequentially deposited over the gate electrode <b>10</b>.

Next, as shown in FIG. 1B, the amorphous silicon layer <b>14</b> and the n+ amorphous silicon layer <b>15</b> are patterned to form a semiconductor pattern. Subsequently, with reference to FIG. 1C, a metal layer is deposited on the n+ amorphous silicon layer <b>15</b> and patterned to form a source electrode <b>16</b> and a drain electrode <b>17</b>. Before the deposition of the metal layer, a natural oxide layer (not shown) formed on the n+ amorphous silicon layer <b>15</b> is removed by a wet etch cleaning process using hydrogen fluoride (HF). Accordingly, the contact resistance between the n+ amorphous silicon layer <b>15</b> and both the source electrode <b>16</b> and the drain electrode <b>17</b> is reduced. After forming the source electrode <b>16</b> and the drain electrode <b>17</b>, an exposed portion of the n+ amorphous silicon layer <b>15</b> is etched using the source and drain electrode <b>16</b> and <b>17</b> as mask.

As shown in FIG. 1D, a passivation layer <b>18</b> is deposited and patterned to have a contact hole <b>19</b> exposing the drain electrode <b>17</b>. Finally, an indium tin oxide (ITO) layer is deposited and patterned to form a pixel electrode <b>20</b>.

However, the conventional method for manufacturing a TFT array panel has many problems.

Impurities, which are generated during semiconductor patterning and HF cleaning processes, may remain on the n+amorphous silicon layer <b>15</b>, and cause disconnections of the source electrode <b>16</b> and the drain electrode <b>17</b>. Further, even with HF cleaning, portions of the natural oxide layer remain. The remaining natural oxide coupled with the impurities, degrades an ohmic contact between the n+ amorphous silicon layer <b>15</b> and both the source electrode <b>16</b> and the drain electrode <b>17</b>, thereby reducing the on current (Ion) of the TFT. Furthermore, pixel electrodes <b>20</b> adjacent to a data line (not shown) interposed therebetween, may be short-circuited when forming the pixel electrodes <b>20</b> due to the ITO residues between the pixel electrodes <b>20</b>. Finally, a photomask misalignment occurring in the photolithography process of forming the amorphous silicon layer <b>14</b> may result in stitch defects because the parasitic electrostatic capacitance (Cgd) between the gate electrode <b>10</b> and the drain electrode <b>17</b> becomes different depending on each photo shot.

SUMMARY OF THE INVENTION

It is therefore an object of the present invention to prevent disconnections of a source electrode and a drain electrode.

It is another object of the present invention to improve an ohmic contact between a semiconductor layer and both a source electrode and a drain electrode.

It is still another object of the present invention to prevent short-circuit between the adjacent pixel electrodes.

It is still yet another object of the present invention to reduce stitch defects.

These and other objects are achieved, according to the present invention, by the following process. After a gate wire is formed, a gate insulating layer, an amorphous silicon layer, a doped amorphous silicon layer and a data metal layer are sequentially deposited in vacuum. The data metal layer is patterned to form a data wire. Next, a doped amorphous silicon layer and an amorphous silicon layer are patterned.

In more detail, a TFT array panel is manufactured by a method including following processes. A gate wire is formed on an insulating substrate, then a gate insulating layer, an amorphous silicon layer and a metal layer are sequentially deposited. The metal layer is patterned to form a data line, a. source electrode and a drain electrode, and the amorphous silicon layer is also patterned. A passivation layer having a contact hole, which exposes a part of the drain electrode, is formed. Finally, a pixel electrode, which is connected to the drain electrode through the contact hole, is formed.

It is preferable that the sequence deposition of the gate insulating layer, the amorphous silicon layer and the metal layer is performed in a vacuum state.

It is possible to deposit a doped amorphous silicon layer after the deposition of the amorphous silicon layer in the sequential deposition process of the amorphous silicon layer and the metal layer, the doped amorphous silicon layer also being patterned when patterning the amorphous silicon layer. After patterning the amorphous silicon layer, the doped amorphous silicon layer is etched by using the data line, the source electrode and the drain electrode as etch stopper.

It is also possible to include the deposition of the doped amorphous silicon layer after the deposition of the amorphous silicon layer in the sequential deposition process of the amorphous silicon layer and the metal layer. After patterning the data line, the source electrode and the drain electrode, the doped amorphous silicon layer is etched by using the data line, the source electrode and the drain electrode as etch stopper. It is preferable that the gate insulating layer, the amorphous silicon layer, the doped amorphous silicon layer and the metal layer are sequentially deposited in vacuum. An equipment that has integrated a sputter equipment and a chemical vapor deposition (CVD) equipment is used for this purpose.

The amorphous silicon layer may be patterned as follows. A photoresist pattern is formed through coating, exposure and development to have a width of 0.1 to 0.4 μm wider than the source electrode and the drain electrode but have the same width as or narrower than the data line. Next, the amorphous silicon layer is overetched to make a groove with a depth of 0.1 to 0.4 μm under the data line.

The gate wire may be a single layer and made of one of Al, an Al alloy, Mo, a Mo alloy, Cr, a Cr alloy, Ta and a Ta alloy, or double-layered and made of any two of the above described materials.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1A to <b>1</b>D are cross-sectional views of a TFT array panel as it undergoes sequential processing steps according to a conventional manufacturing method.

FIGS. 2, <b>4</b>, <b>6</b>, <b>8</b> and <b>10</b> are layout views of a TFT array panel as it undergoes sequential processing steps according to a manufacturing method of a preferred embodiment of the present invention.

FIGS. 3, <b>5</b>, <b>7</b>, <b>9</b> and <b>11</b> are cross-sectional views respectively taken along lines III-III′ of FIG. 2, IV-IV′ of FIG. 4, VI-VI′ of FIG. 6, VIII-VIII′ of FIG. <b>8</b> and XI-XI′ of FIG. <b>10</b>.

FIG. 12 is a diagram showing the organization of the sputter-CVD equipment used in the preferred embodiment of the present invention.

FIG. 13 is a graph comparing an on current of the TFT according to the preferred embodiment of the present invention with that of the conventional TFT.

FIG. 14 is a graph comparing a contact resistance between a doped amorphous silicon layer and both a source electrode and a drain electrode of the TFT according to the preferred embodiment of the present invention with that of the conventional TFT.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention now will be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. In the drawings, the thickness of layers and regions are exaggerated for clarity. Like numerals refer to like elements throughout. It will be understood that when an element such as a layer, region or substrate is referred to as being “on” another element, it can be directly on the other element or intervening elements may also be present. In contrast, when an element is referred to as being “directly on” another element, there are no intervening elements present.

FIGS. 2, <b>4</b>, <b>6</b>, <b>8</b> and <b>10</b> are layout views of a TFT array panel as it undergoes sequential processing steps according to a manufacturing method of a preferred embodiment of the present invention. FIGS. 3, <b>5</b>, <b>7</b>, <b>9</b> and <b>11</b> are cross-sectional views respectively taken along lines III-III′ of FIG. 2, IV-IV′ of FIG. 4, VI-VI′ of FIG. 6, VIII-VIII′ of FIG. <b>8</b> and XI-XI′ of FIG. <b>10</b>.

First, as shown in FIGS. 2 and 3, a first metal layer of Al or an Al alloy such as Al—Nd and a second metal layer of one of Mo, Ta, Cr or their alloys are sequentially deposited on a non-conductive transparent substrate <b>100</b> and patterned by using a first photolithography process to form gate lines <b>210</b> and <b>220</b>, a gate electrode <b>230</b> and gate line bridges <b>240</b> and <b>250</b>. Accordingly, each of the gate lines <b>210</b> and <b>220</b>, the gate electrode <b>230</b> and the gate line bridges <b>240</b> and <b>250</b> has a double-layer structure. That is, each of the gate lines <b>210</b> and <b>220</b> includes a lower metal layer <b>211</b> and an upper metal layer <b>212</b>; the gate electrode <b>230</b> includes a lower metal layer <b>231</b> and an upper metal layer <b>232</b>; and the gate line bridge <b>250</b> includes a lower metal layer <b>251</b> and an upper metal layer <b>252</b>. However, these elements are not limited to a double-layer structure and it is possible to employ a single-layer structure.

Next, as shown in FIGS. 4 and 5, a gate insulating layer <b>300</b>, a amorphous silicon layer <b>400</b> and a doped amorphous silicon layer <b>500</b> are sequentially deposited using a CVD method. Subsequently, the substrate <b>100</b> is moved to a sputter chamber, which is integrally formed with other chambers of CVD equipment, and a Cr layer <b>600</b> is deposited on the doped amorphous silicon layer <b>500</b> by sputtering. At this time, it is preferable that the substrate <b>100</b> is maintained in a vacuum state throughout the entire process of depositing the four layers and not to be exposed to air. This prevents oxidization of an upper surface of the doped amorphous silicon layer (<b>500</b>). This vacuum state is made possible by using a new equipment having a sputter chamber in addition to the conventional CVD equipment.

The four layers may be sequentially deposited without using such a new equipment, but the substrate <b>100</b> is exposed to air while moving it from the CVD equipment to the sputter chamber. This exposure may form an oxide layer on the upper surface of the doped amorphous silicon layer <b>500</b>.

FIG. 12 is a diagram of showing the organization of the CVD equipment that has integrated with a sputter chamber. The CVD equipment includes a load lock chamber in which the substrate <b>100</b> stands by, a preheat chamber that heats the substrate <b>100</b> before deposition, a plurality of process chambers in which thin layers are deposited by a CVD method, and a sputter chamber in which a metal layer is deposited.

The four layers are sequentially deposited using the above described equipment as follows. First, the substrate <b>100</b>, on which the gate wire pattern is formed is placed in the load lock chamber, and then moved into the preheat chamber to be preheated. Next, the substrate <b>100</b> is moved into the first process chamber where the gate insulating layer <b>300</b> and the amorphous silicon layer <b>400</b> are deposited on the substrate <b>100</b>. Following this step, the substrate <b>100</b> moves into the second process chamber where the doped amorphous silicon layer <b>500</b> is deposited. Then, the substrate <b>100</b> moves into the sputter chamber through a vacuum passage where the Cr layer <b>600</b> is doped on the doped amorphous silicon layer <b>500</b>. The vacuum passage can be formed by a passage chamber. At this time, the gate insulating layer <b>300</b>, the amorphous silicon layer <b>400</b> and the doped amorphous silicon layer <b>500</b> are deposited respectively to a thickness of 3,000 to 6,000 Å, 1,000 to 3,000 Å and 200 to 1,000 Å.

As shown in FIGS. 6 and 7, a photoresist layer pattern <b>640</b> is formed by using a second photolithography process, and -the exposed Cr layer <b>600</b> is etched to form a data wire pattern including a data line <b>630</b>, a source electrode <b>610</b> and a drain electrode <b>620</b>.

Next, as shown in FIGS. 8 and 9, the doped amorphous silicon layer <b>500</b> and the amorphous silicon layer <b>400</b> are patterned. There are two methods to pattern the doped amorphous silicon layer <b>500</b> and the amorphous silicon layer <b>400</b>.

In the first method, the doped amorphous silicon layer <b>500</b> is etched to form a pattern by using the photoresist layer <b>640</b> or the data wire pattern as etch stopper. Next, another photoresist layer is coated and patterned to form a photoresist layer pattern (not shown), which covers peripheries of the source electrode <b>610</b> and the drain electrode <b>620</b> to protect the thin film transistor's channel area, by using a third photolithography process. Next, the exposed amorphous silicon layer <b>400</b> is etched.

In the second method, the photoresist layer <b>640</b>, which is used to form the data wire pattern, is first removed. Next, another photoresist layer is coated and patterned to form a photoresist layer pattern (not shown), which covers peripheries of the source electrode <b>610</b> and the drain electrode <b>620</b> in order to protect the channel area of the thin film transistor, by using the third mask. Subsequently, the exposed amorphous silicon layer <b>400</b> and the doped amorphous layer <b>500</b> are simultaneously etched, and then photoresist layer pattern is removed. Finally, the doped amorphous silicon layer <b>500</b> is etched to form the last pattern by using the data wire pattern as etch stopper.

Whichever method is used, the photoresist layer pattern is formed to have an extra width of 0.1 to 0.4 μm wider than the length that completely covers the source electrode <b>610</b> and the drain electrode <b>620</b>, and to have a boundary line that is identical to or narrower than that of the data line under the photoresist layer pattern. Also, the amorphous silicon layer <b>400</b> is overetched to form a groove having a depth of 0.1 to 0.4 μm under the data line <b>630</b>. The doped amorphous silicon layer <b>500</b> may also be overetched to form a groove having a depth of 0.1 to 0.4 μm under the data line <b>630</b>.

Next, as shown in FIGS. 10 and 11, a passivation layer <b>700</b> is deposited and patterned to form a contact hole <b>710</b> using a fourth photolithography process, which exposes the drain electrode. An ITO layer is deposited on the passivation layer <b>700</b> and patterned to form a pixel electrode <b>800</b>, which is connected to the drain electrode <b>620</b> through the contact hole <b>710</b>, using a fifth mask.

In the step of forming the pixel electrode <b>800</b>, even if the ITO layer is not fully etched and leaves residual ITO layer on the passivation layer <b>700</b> over the data line <b>630</b>, the remaining ITO layer is broken off at portions (A), which are grooves formed by the overetched amorphous silicon layer <b>400</b> and the overetched doped amorphous silicon layer <b>500</b>. Therefore, the pixel electrodes <b>800</b>, which are located on opposing sides of the data line <b>630</b>, can be prevented from being short-circuited.

FIG. 13 is a graph comparing an on current of the TFT according to the preferred embodiment of the present invention with that of the conventional TFT. FIG. 14 is a graph comparing a contact resistance between the doped amorphous silicon layer <b>500</b> and both the source electrode <b>610</b> and the drain electrode <b>620</b> of the TFT according to the preferred embodiment of the present invention with that of the conventional TFT.

In FIG. 13, the on current distribution of the conventional TFT is represented by white triangles, and that of the TFT according to the preferred embodiment of the present invention is represented by black triangles.

As shown in FIG. 13, regardless of what value a gate voltage (Vg) and a data voltage are, the on current of the TFT according to the present invention is lager than that of the conventional TFT. This is because of mobility differences. That is, the mobility of the conventional TFT is about 0.5 cm²/v·sec, but that of the TFT according to the present invention is about 0.79 cm²/v·sec.

In FIG. 14, the distribution of the contact resistance according to the voltage variance between the source electrode and the drain electrode of the TFT according to the present invention is represented by black circles, and that of the conventional TFT is represented by white circles.

As shown in FIG. 14, regardless of the level of a voltage (Vds) between the source electrode and the drain electrode, the contact resistance of the TFT according to the present invention is smaller than that of the conventional TFT.

In the present invention described above, an oxide layer is prevented by sequentially depositing layers in a vacuum state. Therefore, the on current of the TFT is increased. Further, HF cleaning can be omitted because no oxide layer is formed. Therefore, the overall TFT manufacturing process is simplified.

Moreover, adjacent pixel electrodes is protected from short-circuit by forming grooves between the pixel electrodes. Since the semiconductor layer is widely distributed under the source electrode and the drain electrode, even if the photomask misaligns while forming the semiconductor layer pattern and the source and the drain electrodes, the parasitic capacitance between the gate electrode and the source electrode does not substantially fluctuate. Therefore, kickback voltage also does not widely fluctuate. This ultimately prevents the stitch defects.

In the drawings and specification, there have been disclosed typical preferred embodiments of the present invention and, although specific terms are employed, they are used in a generic and descriptive sense only and not for purposes of limitation, the scope of the invention being set forth in the following claims.